{"id":44642818,"url":"https://github.com/bytelogiccore-spec/dbx","last_synced_at":"2026-04-03T01:10:51.907Z","repository":{"id":338640286,"uuid":"1157884035","full_name":"bytelogiccore-spec/DBX","owner":"bytelogiccore-spec","description":"High-performance embedded database with 5-Tier Hybrid Storage. 29x faster than SQLite • Pure Rust • GPU-Accelerated • MVCC Transactions","archived":false,"fork":false,"pushed_at":"2026-02-17T21:34:01.000Z","size":91441,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-02-18T01:21:10.661Z","etag":null,"topics":["database","embedded-database","gpu","high-performance","htap","key-value-store","mvcc","rust","sql"],"latest_commit_sha":null,"homepage":"https://bytelogiccore-spec.github.io/DBX/","language":"Rust","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/bytelogiccore-spec.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":null,"funding":".github/FUNDING.yml","license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null},"funding":{"github":["ByteLogicStudio"],"ko_fi":"Q5Q41TDHWG"}},"created_at":"2026-02-14T13:03:50.000Z","updated_at":"2026-02-17T21:34:05.000Z","dependencies_parsed_at":null,"dependency_job_id":"57447080-8391-4deb-9150-ef65365585fc","html_url":"https://github.com/bytelogiccore-spec/DBX","commit_stats":null,"previous_names":["bytelogiccore-spec/dbx"],"tags_count":5,"template":false,"template_full_name":null,"purl":"pkg:github/bytelogiccore-spec/DBX","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bytelogiccore-spec%2FDBX","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bytelogiccore-spec%2FDBX/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bytelogiccore-spec%2FDBX/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bytelogiccore-spec%2FDBX/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/bytelogiccore-spec","download_url":"https://codeload.github.com/bytelogiccore-spec/DBX/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bytelogiccore-spec%2FDBX/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29597852,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-02-18T22:25:43.180Z","status":"ssl_error","status_checked_at":"2026-02-18T22:25:42.766Z","response_time":162,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["database","embedded-database","gpu","high-performance","htap","key-value-store","mvcc","rust","sql"],"created_at":"2026-02-14T19:06:27.610Z","updated_at":"2026-04-03T01:10:51.896Z","avatar_url":"https://github.com/bytelogiccore-spec.png","language":"Rust","funding_links":["https://github.com/sponsors/ByteLogicStudio","https://ko-fi.com/Q5Q41TDHWG","https://ko-fi.com/YOUR_KO_FI_LINK"],"categories":[],"sub_categories":[],"readme":"# DBX — Next-Generation HTAP Embedded Database\n{: .fs-9 }\n\n[![Version](https://img.shields.io/badge/version-0.2.0--beta-blue.svg)](https://github.com/ByteLogicCore/DBX)\n[![License](https://img.shields.io/badge/license-MIT-green.svg)](LICENSE)\n[![Rust](https://img.shields.io/badge/rust-2024-orange.svg)](https://www.rust-lang.org)\n[![Documentation](https://img.shields.io/badge/docs-GitHub%20Pages-blue)](https://bytelogiccore-spec.github.io/DBX/)\n\n### 💖 Support This Project\nIf you find DBX useful, please consider supporting its development!\n\n[\u003cimg src=\"https://ko-fi.com/img/githubbutton_sm.svg\" alt=\"ko-fi\" height=\"36\"\u003e](https://ko-fi.com/YOUR_KO_FI_LINK)\n\nYour support helps with:\n- 🚀 New features and performance optimizations\n- 🐛 Bug fixes and stability improvements\n- 📚 Documentation and tutorials\n- 💻 Test infrastructure and CI/CD maintenance\n\n---\n\n**DBX** is a high-performance, embedded database designed for modern **HTAP (Hybrid Transactional/Analytical Processing)** workloads. Built with a unique **5-Tier Hybrid Storage** architecture, it bridges the gap between ultra-fast in-memory transactions and massive-scale columnar analytics.\n{: .fs-6 .fw-300 }\n\n---\n\n## ⚡ The DBX Edge: Why Choose Us?\n\n### 🚄 1. Infinite Scalability (5-Tier Architecture)\nUnlike traditional databases that force a choice between speed and capacity, DBX flows data through 5 specialized tiers:\n- **Tier 1 (Delta)**: Ultra-fast BTreeMap for sub-millisecond writes.\n- **Tier 2 (Cache)**: Apache Arrow-based columnar cache for instant OLAP.\n- **Tier 3 (WOS)**: SSD-optimized MVCC storage for snapshot isolation.\n- **Tier 4 (Index)**: High-speed Bloom filters for near-zero latency probes.\n- **Tier 5 (ROS)**: Compact Parquet storage for petabyte-scale archiving.\n\n### 🏎️ 2. Blazing Fast Performance\nBenchmark results (10,000 records) show DBX outperforming industry standards with the new **Fast-Path** optimization (v0.2.0):\n- **Local Scan**: **51µs (microseconds)** — Ultra-low latency via local bypass 🔥\n- **Memory INSERT**: **1.16x faster** than SQLite (25ms vs 29ms) ✅\n- **Analytics (OLAP)**: **Up to 29x faster** than traditional B-Tree engines.\n\n### 🧠 3. GPU-Native Analytics\nDBX is the first embedded database to offer **first-class CUDA acceleration**.\n- **4.5x faster** filtering and aggregation on large datasets.\n- Seamlessly offload heavy JOINs and GROUP BY operations to the GPU.\n\n### ⚡ Zero-Lock Concurrency (MVCC) + Distributed Locks (DLM)\nReaders never block writers, and writers never block readers.\nEquipped with a **Network-Aware Distributed Lock Manager (DLM)** that uses Fencing Tokens and Adaptive Leases via `GridDatabaseAsync`, while strictly preserving single-node HTAP latency.\n- Built-in absolute **Atomic CAS operations** (`insert_if_not_exists`, `compare_and_swap`).\n- Super-fast 1024-Striped **Row-level Latch Lock Manager**, removing heavy table mutexes from the main data store. (Providing blazing-fast concurrent operations across 5 tiers without table-level bottlenecks).\n  *(Disclaimer: Secondary Index updates currently utilize standard `RwLock` for tree integrity, resulting in brief column-level locks during write-heavy index operations.)*\n\n### 🌐 5. Distributed Grid Architecture (Grid Engine)\nA **Data Grid** is a distributed computing architecture where multiple machines connected over a network act as **one unified database**. DBX has this grid engine built-in, allowing you to start as a single-node embedded DB and — without any code changes — scale horizontally across networked nodes.\n\n\u003e **💡 Why Grid?** Traditional embedded databases (SQLite, Sled, etc.) are confined to a single process. DBX is the **only embedded database** capable of real-time data replication and distribution across nodes over the network.\n\n- **Network Replication (QUIC Transport)**: Ultra-low latency data replication between nodes via TLS 1.3 QUIC protocol. Zero Head-of-Line Blocking ensures stability even in large clusters.\n- **Quorum-based High Availability**: Raft-like leader election with majority ACK approval ensures data consistency and service continuity even when some nodes go down.\n- **Multi-Master Failover**: Automatic leader election on failure with Vector Clock-based conflict resolution to maintain data integrity.\n- **Auto-Sharding (Cross-Node)**: Data is automatically distributed across nodes. Adding a node triggers automatic rebalancing, scaling capacity and throughput linearly.\n- **Distributed Transactions (2PC)**: Transactions spanning multiple nodes are guaranteed atomic via Two-Phase Commit.\n\n---\n\n## 📦 Modern Capabilities\n\n### 💎 Materialized Views\nDefine complex analytical queries and let DBX handle the heavy lifting.\n- **Auto-Refresh**: Background threads keep your results fresh every 60 seconds.\n- **Transparent Caching**: SQL queries hit the cache automatically for instant responses.\n\n### 🌊 Real-time Streaming (CDC)\nBuilt-in `StreamIngester` for high-throughput data pipelines.\n- **MPSC Pipeline**: Concurrently ingest from thousands of producers.\n- **Full DML Support**: Real-time INSERT, UPDATE, and DELETE processing.\n\n---\n\n## 🦀 Rust Ecosystem Compatibility\n\nDBX is designed to integrate seamlessly with the modern Rust data ecosystem.\n- **Native Serde Support**: Store and retrieve custom Rust `struct`s directly without manual byte conversions via the `DatabaseSerde` trait.\n- **Async First Driver**: A non-blocking wrapper `DatabaseAsync` seamlessly offloads heavy I/O to `tokio::task::spawn_blocking`, making it perfect for heavily concurrent async web servers.\n\n---\n\n## 🏗️ Architecture Visualization\n\n```mermaid\ngraph TD\n    A[Application] --\u003e|Write| B[Tier 1: Delta Store]\n    B --\u003e|Flush| C[Tier 2: Columnar Cache]\n    C --\u003e|Persist| D[Tier 3: WOS - SSD]\n    D --\u003e|Compact| E[Tier 4: Index]\n    E --\u003e|Archive| F[Tier 5: ROS - Parquet]\n    \n    G[GPU Acceleration] -.-\u003e|Speed Up| C\n    G -.-\u003e|Speed Up| F\n    \n    style B fill:#f9f,stroke:#333,stroke-width:2px\n    style C fill:#bbf,stroke:#333,stroke-width:2px\n    style F fill:#bfb,stroke:#333,stroke-width:2px\n```\n\n---\n\n## 🚀 Quick Start (Rust)\n\n```rust\nuse dbx_core::Database;\n\nfn main() -\u003e dbx_core::DbxResult\u003c()\u003e {\n    // Open a database in memory or on disk\n    let db = Database::open_in_memory()?;\n\n    // Lightning-fast CRUD\n    db.insert(\"users\", b\"user:123\", b\"{\\\"name\\\": \\\"Alice\\\", \\\"age\\\": 30}\")?;\n    let val = db.get(\"users\", b\"user:123\")?;\n\n    // Powerful HTAP SQL\n    let results = db.execute_sql(\"SELECT name, AVG(age) FROM users GROUP BY name\")?;\n\n    Ok(())\n}\n```\n\n---\n\n## 🤝 Support \u0026 Contributing\n\nDBX is an open-source project powered by the community.\n- ⭐️ **Star us** on GitHub to show your support!\n- 🐛 **Report issues** or request features via GitHub Issues.\n- 🛠️ **Contribute**: Check our [Contributing Guide](./legal/english/CONTRIBUTING.md) for build optimizations (LLD, mold, etc.).\n\n---\n\n**Made with ❤️ in Rust for the Future of Data.**\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbytelogiccore-spec%2Fdbx","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fbytelogiccore-spec%2Fdbx","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbytelogiccore-spec%2Fdbx/lists"}